Edited by: Laura Ferraiuolo, The University of Sheffield, United Kingdom
Reviewed by: Varun Kesherwani, University of Nebraska Medical Center, United States; Ru-Ping Dai, Central South University, China
†These authors have contributed equally to this work
This article was submitted to Neurodegeneration, a section of the journal Frontiers in Neuroscience
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Postoperative cognitive dysfunction (POCD) is serious disorder in the central nervous system common in aged patients after anesthesia. Although its clinical symptoms are well recognized, however, the molecular etiology of the POCD remains unrevealed. Similarly, neither gold standard molecular diagnosis nor effective treatment is available for POCD until the present. Therefore, we aimed to explore the molecular mechanism of this disorder through investigating lncRNAs and mRNAs associated with POCD human patients and investigate their underlying regulatory pathways. In this study, we recruited 200 patients requiring hip or knee replacement surgery. Their neurological functions were assessed at two time points, 1 day before the surgery and 30 days post-surgery. In parallel, serum samples were collected from the participants to analyze lncRNAs and mRNAs differential expression profile between POCD and non-POCD patients using microarray analysis. To further investigate the role differentially expressed mRNA and lncRNAs, Gene Ontology (GO), pathway analyses on mRNAs and lncRNA-mRNA interaction network were performed. As a result, 68 lncRNAs and 115 mRNAs were dysregulated in the POCD group compared to non-POCD group. Among them, the top 10 upregulated lncRNAs and 10 downregulated lncRNAs were listed for enrichment analysis. Interestingly, we found that these lncRNA and mRNA are involved in biological process, molecular function, and cellular component in addition to various signaling pathways, suggesting that the pathogenesis of POCD involves lncRNAs and mRNAs differential expression. Consequently, the genetic dysregulation between the non-POCD and POCD patients participates in the occurrence and development of POCD, and could be served as diagnostic biomarkers and drug targets for POCD treatment.
Postoperative cognitive dysfunction (POCD) is a severe central nervous system (CNS) complication in the cognitive function common in elderly patients after anesthesia and surgery. The condition involves symptoms rang from mild to serious cognitive deficits affecting life quality (
Long non-coding RNAs (lncRNAs) are set of non-coding RNA with more than 200 nucleotides lengths (
We selected hip and knee replacement patients as clinical cases to detect the lncRNAs and mRNAs profiles in POCD patients non-POCD patients using microarray analysis. Furthermore, Gene Ontology (GO) and KEGG analyses were performed to study the potential functions of the dysregulated lncRNAs and mRNA. Finally, network interaction of lncRNAs–mRNAs in the POCD patients was constructed.
The study was approved by the ethic committee of Affiliated Traditional Chinese Medicine Hospital of Southwest Medical University (project number KY2018001) and registered with Chinese Ethics Committee of Registering Clinical Trials in 2018 (Unique identifier ChiCTR-INR-15007607). All participating patients were informed with the aim of the study and signed the related consent forms.
Two hundred patients admitted to Affiliated Traditional Chinese Medicine Hospital of Southwest Medical University to carry out hip or knee replacement surgery have participated in this study. Patients selected based on firm criteria which were (1) age at >60 years old; (2) anesthetic procedure (American Society of Anesthesiologists Grade 1–3); and (3) mental health. It was decided to exclude patients beyond 85 years old, patients with nervous system diseases, preoperative stoke, and other central nerve system diseases cases (preoperative MMSE < 24 scores), and patients who suffered from multiple injuries and craniocerebral injury.
All patients were asked to fast from food 8 h before the operation and from water 3 h before anesthesia. The anesthetic procedure was combined of spinal and epidural anesthesia at the L3–L4 intervertebral disc in a side-lying position using an 18-gage needle and 25-gage spinal needle (Tuo Ren, Henan, China). Then, the 2 ml mixed solution consisting of 0.5% bupivacaine (Hua Lu, Shandong, China) and 3.33% glucose (Shuang He, Beijing, China) was slowly injected to subarachnoid space. The vital sign of patients was monitored during the surgery. All the participants were given the same analgesia doses 12 μg/kg fentanyl (Ren Fu, Yichang, China) and 10 mg tropisetron hydrochloride (Ren Fu, Yichang, China) for 2 days.
For cognitive function evaluation, a battery of neuropsychological tests were followed 1 day before and 1 month after the surgery by trained physician. The tests consisted of the MMSE, Symbol Digit Modalities Test (SDMT), Digit Span Test (DST), Trail Making Test A, and Clock Drawing Test (CDT). The cognitive functions were principally assessed including memory, learning concentration, and executive function. Briefly, MMSE was designed to evaluate and quantify the global cognitive state, assess memory, spatial–temporal orientation, and reasoning (
A group 30 health volunteers were also assessed as normal reference of cognitive functions and were corresponding with their patient’s counterparts based on education level, age, and gender. The cognitive functions were evaluated twice in 7 days intervals. Patients were separated into POCD and non-POCD groups. The (SD) value, mean variation, and baseline score were calculated to statistically validate the data (
Three days after the surgery, 8 ml of peripheral blood samples were collected from the patients in pro-coagulant tube and centrifuged (at speed of 2500 rpm for 10 min). The supernatant was taken and stored in liquid nitrogen until the test. Random number table was applied to select three patients with early POCD (labeled A, B, C) and three patients with non-POCD was as control (labeled D, E, F). Total RNA was extracted using TRIzol reagent kit based on vendor instructions, the plasma. RNA integrity was assessed by Agilent Bioanalyzer 2100 (Agilent Technologies), and the purity was measured with NanoDrop2000 spectrophotometer (Thermo Scientific). RNA samples with OD260/280 > 1.8 and OD260/230 > 1.5 were taken for further testing.
RNA samples were sent to Shanghai OE Biotech Inc. (Shanghai, China) to carry out. The sample labeling, hybridization, and washing were performed according to service provider protocols. In brief, cDNAs was synthesized for from RNA samples then synthesized cRNAs. Second cycle cDNAs were synthesized from cRNAs and followed by fragmentation and biotin labeling, the second cycle cDNAs were hybridized onto the Affymetrix Human OE lncRNA microarray (Affymetrix, Santa Clara, CA, United States), comprising 66,741 human lncRNAs and 25,986 coding mRNA. The chips were then washed, stained, and scanned by Affymetrix Scanner 3000 (Affymetrix, Santa Clara, CA, United States).
Affymetrix Gene Chip Command Console (version 4.0, Affymetrix) software was used to extract raw data. Expression Console (version1.3.1, Affymetrix) software offered RMA normalization for both gene and exon level analysis. Then, Genesrping software (version 13.1; Agilent Technologies) was employed to carry out the basic analysis. Venn Diagram software was performed to generate high-resolution Venn plots.
To understand the function of lncRNAs associated with the PODC, we first established the role of differentially expressed mRNA through GO and KEGG analyses. The lncRNAs then were linked to their adjacent mRNAs to establish their potential function (
Samples self-test.
In this study, SPSS software version 17.0 (SPSS, Inc., Chicago, IL, United States) and GraphPad Prism 5.0 (GraphPad Software, Inc., La Jolla, CA, United States) were used for data analysis and the experimental results were presented as mean values ± standard deviation (SD);
Of these 200 patients, 181 patients (91.5%) achieved the whole tests, while 19 patients (9.5%) did not complete the tests due personal reasons. The general characteristics of the patients including age, body mass index, gender, and education are presented in
General characteristics and cognitive functions of participants from study and control groups at baseline
Item (baseline) | Study group ( |
Control group ( |
|
---|---|---|---|
Age (years) | 77.1 ± 2.5 | 77.3 ± 3.1 | 0.696 |
Weight (kg) | 63.8 ± 10.7 | 65. 7 ± 12.4 | 0.380 |
Education (years) | 8.3 ± 2.4 | 9.1 ± 2.6 | 0.096 |
MMSE scores | 28.2 ± 1.6 | 28.8 ± 2.1 | 0.071 |
DST scores | 16.3 ± 1.8 | 16.6 ± 1.6 | 0.392 |
CDT scores | 2.5 ± 0.5 | 2.6 ± 0.6 | 0.326 |
SDMT scores | 31.3 ± 3.2 | 30.4 ± 3.5 | 0.161 |
TMT-A scores | 41.1 ± 5.8 | 42.4 ± 6.6 | 0.266 |
General characteristics, intraoperative indicators and cognitive functions of patients from non-POCD and POCD at baseline.
Item (baseline) | Non-POCD ( |
POCD ( |
|
---|---|---|---|
Age (years) | 76.8 ± 2.7 | 77.2 ± 3.1 | 0.531 |
Weight (kg) | 63.4 ± 11.2 | 64 ± 12.4 | 0.820 |
Education (years) | 8.3 ± 1.4 | 8.6 ± 2.6 | 0.357 |
ASA classification (I/II/III) | 32/104/24 | 5/14/2 | 0.501 |
Operation time (min) | 84.9 ± 18.8 | 79.8 ± 19.6 | 0.246 |
Intraoperative blood loss (ml) | 296.4 ± 37.2 | 290.6 ± 43.8 | 0.512 |
Intraoperative urine output (ml) | 376.3 ± 86.2 | 370.4 ± 84.5 | 0.768 |
Intraoperative blood sugar (mmol/l) | 5.71 ± 0.58 | 5.78 ± 0.66 | 0.610 |
MMSE scores | 28.0 ± 1.4 | 28.4 ± 1.8 | 0.236 |
DST scores | 16.1 ± 1.4 | 16.4 ± 1.5 | 0.361 |
CDT scores | 2.4 ± 0.5 | 2.7 ± 0.6 | 0.013 |
SDMT scores | 31.0 ± 3.6 | 31.6 ± 3.9 | 0.478 |
TMT-A scores | 40.1 ± 7.8 | 42.3 ± 8.6 | 0.231 |
Neuropsychological test scores of non-POCD group and POCD group.
Item | Non-POCD ( |
POCD ( |
|
---|---|---|---|
MMSE scores | 27.2 ± 0.6 | 25.9 ± 1.1 | 0.000 |
DST scores | 14.1 ± 1.6 | 13.1 ± 1.1 | 0.006 |
CDT scores | 2.3 ± 0.5 | 1.9 ± 0.7 | 0.001 |
SDMT scores | 31.3 ± 3.2 | 28.9 ± 3.5 | 0.002 |
TMT-A scores | 41.4 ± 5.1 | 38.7 ± 6.1 | 0.027 |
Chip data dispersion was illustrated in box–whisker plot, scatter plot, and PCA. From box–whisker plot, no statistical variance was seen neither in the lncRNAs distributions nor in mRNAs expression profile in the samples (
Hierarchical clustering analysis data and heatmap (
Differential expression of lncRNAs.
Gene Ontology analysis for the main overexpressed mRNA has revealed that these transcriptomes are involved in biological processes through series of functions including deacetylation, cytoplasmic mRNA processing, stem cell population maintenance, and positive regulation of immunity against viruses by vesicle organization, regulation of mitotic cell cycle, peptidyl-threonine phosphorylation, endocytosis, histone deacetylation, and actin cytoskeleton reorganization (
Analysis of bioinformatics for the upregulated genes.
On the basis of the enrichment scores, the top 10 biological processes targeted by downregulated mRNAs were identified as cytoplasmic mRNA processing body assembly, snRNA transcription, polymerase II promoter, cell division, regulation of mitotic cell cycle, outflow tract morphogenesis, stem cell population maintenance, regulation of translation, regulation of fertilization, meiotic chromosome condensation, and regulation of myotube differentiation (
Analysis of bioinformatics for the downregulated genes.
Based on the information of
Analysis of upregulated lncRNAs-mRNA co-expression.
In addition, the top 10 downregulated lncRNAs were also associated with many mRNA sequences. As the most downregulated lncRNA in POCD, NR_024187 was correlated with that of NEK3, XKRX, VNN1, TBXAS1, and HDAC5 (
Analysis of downregulated lncRNAs-mRNA co-expression.
Microarray data analysis also showed the overexpressed genes. Total of 115 mRNA sequences were dysregulated between the POCD and control groups (FC > 2,
Differential expression of mRNAs. Heatmaps of mRNAs in POCD patients vs. non-POCD patients.
Postoperative cognitive dysfunction is a postoperative complication with high incidence and poor prognosis (
As a class of important ncRNA, lncRNA plays vital role in central system (
In the present study, various neuropsychological tests were adopted to evaluate cognitive functions 1 day before and 30 days after knee or hip replacement surgery to distinguish between POCD and non-POCD patients. Then, lncRNA sequences were studied from all the patients using developed microarray chips. Microarray analyses revealed that there were total of 29 upregulated lncRNAs and 39 mRNAs, 39 downregulated lncRNAs, and 76 mRNAs associated with POCD. Likewise, the GO and KEGG were performed to predict the roles of the dysregulated lncRNAs, and the co-expression mRNA. Furthermore, network interaction of lncRNAs and mRNAs in the POCD patients was constructed to explore the relationship between lncRNAs and mRNAs with the disease. As POCD symptoms are memory decline, poor attention, learning disability, and deficit function execution. Therefore, this experiment uses a series of neuropsychology tests to assess these five categories of cognition functions to distinguish whether cognitive impairment existed before surgery it has occurred after the surgery. To increase the accuracy, all cognitive functions were measured by more than one test to diagnose the POCD patients.
Bioinformatics analyses were aimed to indicate the potential functions of the dysregulated lncRNAs that have been identified in our study. GO analysis has illuminated the functions of genes through biological process, cellular component, and molecular function. While KEGG is normally used to study the possible roles of expressed RNA with correlation to cell signaling pathways (
In our analysis of biological process, it was noticed that the differentially expressed lncRNAs, mainly participated in process of protein and histone deacetylation, cytoplasmic mRNA processing body assembly, regulation of cell cycle, and stem cell population. Previous study has also pointed out the role of histone acetylation/deacetylation in neurodegenerative conditions (
Furthermore, pathway analysis revealed that there were 10 pathways implicated in POCD patients, Epstein–Barr virus infection, and pantothenate and CoA biosynthesis were the highly enriched pathways in POCD patients. Epstein–Barr virus, a human gamma-herpesvirus, potentially facilitates the damage of autoimmunity and CNS tissue, causing psychiatric symptoms and cognitive dysfunction (
Until the present, the functions of lncRNAs are not entirely clear. The co-expression network of aberrantly expressed mRNAs and lncRNAs in POCD was constructed to further understand of the role of lncRNAs via calculation and inference.
In the present study, we have selected the top 10 up and downregulated lncRNAs to construct the lncRNA-mRNA co-expression network of interactions to predict the genes targeted by lncRNA. We found that the increased lncRNA, lnc-FAM53B-2_1, correlated with that of SMARCE1, ATXN2, SMC4, and IL23R, while main reduced lncRNA was NR_024187 and has influence on NEK3, XKRX, VNN1, TBXAS1, and HDAC5. Previous study showed SMARCE1 to be associated with the occurrence and development of non-syndromic intellectual disability (
However, it will be worthwhile to evaluated neurological function in long-term manner in POCD patients. In addition, more hands on researches should be conducted to investigate the exact mechanisms and function of genes that have been reported in this study to examine the specific interaction mechanisms between lncRNAs and POCD.
We have recorded that lncRNAs and mRNAs in serum were differentially expressed in patients with POCD compared with patients without POCD after hip or knee replacement surgery. These vital findings may contribute to well understand the mechanism of POCD and provide new biomarkers for diagnosis and monitoring the progression of POCD and potential therapeutic target for POCD treatment.
F-XY and YZ conceived and designed the experiments. YZ, Y-XL, Q-XX, and QL performed the experiments. Y-XL, RD, ID, and JB analyzed the data. YZ, MA-H, and Y-XL wrote the manuscript. MA-H and ID participated in improving the paper and making the language more authentic during the revision.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
We greatly appreciate professor Ting-Hua Wang for his faithful contribution in modifying the paper.
The Supplementary Material for this article can be found online at: